Sequential Procedures for Detecting Parameter Changes in a Time Series Model,
Abstract
Procedures are proposed for monitoring forecast errors in order to detect changes in a time series model. These procedures are based on likelihood ratio statistics which consist of cumulative sums. Both a Wald type sequential scheme and an extension of Page's method are considered. The distributional properties of the statistics are approximated under the assumption that the series follows an integrated autoregressive moving average model. The approximation is based on the limiting Wiener process. An example is also given. (Author)
Document Details
- Document Type
- Technical Report
- Publication Date
- Sep 01, 1975
- Accession Number
- ADA027723
Entities
People
- Michael Bagshaw
- Richard A. Johnson
Organizations
- University of Wisconsin–Madison